The Marginal Economic Value of Streamflow from National Forests:
Evidence from Western Water Markets
Thomas C. Brown
U.S. Forest Service, Rocky Mountain Research Station, Fort Collins, Colorado
Evidence from over 2,000 water market transactions that occurred in the western U.S. over the past 14
years (1990 through 2003) was examined to learn who is selling to whom and for what purpose, how much
water is involved, and how much it is selling for. Roughly half of the transactions were sales of water rights;
the rest were water leases. The transactions show that the price of water is highly variable both within and
between western states, reflecting the localized nature of the factors that affect water prices. Ideally, if water
market prices or valuation studies are to be used to help determine the marginal value of water from specific
areas such as national forests, information from local markets or local studies should be used. Lacking sitespecific value information, only rough estimates are possible.
Keywords: water value, streamflow, economic value, water markets
INTRODUCTION
The economic value of a good or service is indicated by
a willingness to sacrifice other goods and services in order
to obtain or retain it, and is typically measured in money
terms, usually as willingness to pay (WTP). WTP may be
determined for changes in the quantity, quality, or timing
of water. This paper deals with the value associated with
water quantity, specifically the quantity of streamflow.
The value of a small change is what economists call
marginal value. Estimates of marginal value are useful
in analyzing policies that cause relatively small changes,
such as changes in streamflow resulting from vegetation
management. Of course, even small changes in streamflow
may have numerous additive downstream effects. For
example, an acre-foot of streamflow increase may first be
used by recreationists, next pass through a hydroelectric
plant, then be diverted to a farm, and finally be diverted
to a city, all the while helping to dilute wastes and enhance
fish habitat. The aggregate value of a change in streamflow
is equal to the sum of its values in the different
instream and offstream uses to which the water is put
during its journey to the sea (e.g., Brown et al. 1990).
For the general case, the aggregate value of a small
change in streamflow (V*) is given by:
M Furniss, C Clifton, and K Ronnenberg, eds., 2007. Advancing
the Fundamental Sciences: Proceedings of the Forest Service National
Earth Sciences Conference, San Diego, CA, 18-22 October 2004, PNWGTR-689, Portland, OR: U.S. Department of Agriculture, Forest
Service, Pacific Northwest Research Station.
where i indicates a water diversion or instream use location;
αi is the proportion of the marginal acre-foot that reaches
a use i (water that is not consumed upstream) (α ≤ 1); βi
is the proportion of the marginal acre-foot reaching use i
that arrives when it actually can be of use (β ≤ 1); and Vi
is the value of the marginal acre-foot in use i assuming the
water is put to use (V ≥ 0).
The task of estimating the marginal value of streamflow
from a national forest thus consists of estimating the Vs
and their respective αs and βs. Our focus here is on Vi.
There are two basic approaches to estimating the
marginal value of water: employing economic valuation
methods and observing water market prices. The suite
of valuation methods (Gibbons 1986; Young 1996)
was developed because markets for goods like water
were uncommon and, when present, rarely competitive.
However, water market activity has increased in recent
years, raising hopes that water markets can offer useful
estimates of water value. This paper presents evidence from
water market transactions in the western United States.
To review how price is related to marginal value,
consider the case of offstream uses depicted in Figure 1. If
demand (i.e., marginal WTP) for water at a point of use
is represented by D, Q1 is the total quantity of streamflow
available, and P1 is the cost of transporting (e.g., pumping)
the diverted water to the point of use, then users desire
to divert Q2 units. Here the net marginal value of the
diverted streamflow (V) is P1-P1 = 0 per unit. Alternatively,
if water supply were constrained at Q3, Q3 units would
be diverted. At a diversion of Q3 units, the marginal value
of delivered water is P2, and V = P2-P1. If a competitive
market for streamflow existed in this location, the market
price would be P1, and if a competitive market for delivered
water existed its market price would be P2.
BROWN
This example thus illustrates two points: (1) streamflow
has value at the margin only when there is not enough
of it to meet all demands; and (2) if the price includes
consideration for storage or delivery, it may overstate the
marginal value of streamflow.
Figure 1. Marginal value of offstream diversion.
WATER MARKETS
Water has become scarcer as population, economic
growth, and changing values in the West have increased
demand for water (Gillilan and Brown 1997). Where
institutions allowed it and transaction costs were not
excessive, the growing scarcity often brought willing buyers
and sellers together in what is called a water market. The
term “water market” lacks a precise definition, but once
a few voluntary trades of water of relatively common
physical and legal characteristics occur, it is said that a
water market has developed.
Water markets require a well-administered system
of transferable water rights. The doctrine of prior
appropriation that underlies water law across the West
allows for clearly defined and transferable water rights, and
state agencies or the courts administer and enforce those
rights, although the states differ in how they implement the
doctrine and administer the water rights systems (National
Research Council 1992).
Water market activity may be limited by physical and
legal constraints. Physical constraints on water trades, such
as uneven water availability and lack of access, are eased
by such structures as diversion dams, canals, and storage
reservoirs, which extend spatial and temporal control over
water delivery. Legal constraints on water sales often exist,
perhaps due to state law (e.g., constraints on moving
water to another basin, or constraints on the availability of
instream flow rights) or federal guidelines for specific water
development projects. As scarcity has intensified, some
constraints on water markets are being loosened (North
1990; Loomis 1992).
459
If water in a water short area were freely traded in an
efficient market, water would be reallocated via trades to
the point where each user was consuming at the point
where the marginal values in all uses were identical (e.g.,
the marginal value in irrigation would be equal to the
marginal value in municipal use or in instream recreation).
In this ideal world, a single market price would emerge
that would indicate the marginal value of raw water in
that market area. However, in the real world, even in
those locations where water markets exist, water rarely
trades so easily or completely. Two reasons for this are
lack of a homogeneous product and lack of market
competitiveness.
Lack of homogeneity is a natural consequence of how the
prior appropriation doctrine accommodates the stochastic
nature of streamflow. The doctrine deals with shortage
by assigning priorities to water rights and temporarily
canceling permission to divert based on those priorities,
beginning with the most junior right and moving as far up
the list of priorities as needed to assure delivery to more
senior rights. Each individual right may have a unique
priority date. Senior rights are worth more than junior
rights because senior rights face less risk of shortage. If each
right is unique, homogeneity of product is compromised.
However, within the overall structure of prior appropriation
there exists a quite different approach know as fractional
flow rights (Eheart and Lyon 1983). With such rights, all
users have equal priority, and shortage is accommodated
in a given time period by lowering the allowable diversion
for all users. The use of fractional flow rights is common
in mutual ditch companies and some water conservancy
districts, wherein water is owned as shares of the total
amount available (Hartman and Seastone 1970). Within
such an organization all members essentially have the
same priority, and the effect of a flow increase available
to the organization is distributed to the members in
proportion to the number of shares each owns, thus
providing homogeneity of product. Many of the more
active water markets deal in shares of such a company or
district.
A fundamental tenet of neoclassical economic theory
is that competitive markets yield prices that reflect the
true marginal economic value of the good being traded.
Competitive markets have many buyers and sellers, do
not artificially restrict price or ability to trade, have low
transaction costs, allow an easy flow of information about
prices and potential trades, and internalize all relevant costs
and benefits of the transaction. Water markets typically
fall short on one or more of these requirements. Many
markets areas are so small that sellers and buyers are few. In
others, laws, regulations, or customs limit price. In many
water markets transaction costs are substantial, involving
460
MARGINAL ECONOMIC VALUE OF STREAMFLOW
administrative and legal requirements. In many markets
information is not readily available. And externalities
(effects on individuals not party to the exchange) commonly
exist, especially in the form of changes in water quality
and instream flow (Howe et al. 1986; Saliba 1987).
Some of these restrictions on the competitiveness of the
market (e.g., a limited number of sellers) may elevate the
price relative to the price that would be established in a
purely competitive market, whereas others tend to depress
the price (e.g., government subsidies, transaction costs,
regulations or customs). Many of the restrictions, such as
transaction costs, will also tend to limit the number of
trades.
Studies of water markets have usually focused in detail
on one or a few specific markets (e.g., Hartman and
Seastone 1970; Saliba et al. 1987; Michelsen 1994; Howe
and Goemans 2003). Only with a detailed examination can
the numerous characteristics of the individual markets be
given their due consideration. This study, in contrast, takes
a broad look across the western United States, emphasizing
geographical scope rather than in-depth focus (see also
Brown 2006). This “big picture” approach offers a look
at how water prices in general have changed over the past
few years and how they differ across locations or across the
purposes for which the water was purchased.
When water is sold in the West, either a water right
changes hands or use of the right is essentially leased for a
period of time. Ownership of a water right conveys access
to a specified quantity of water in perpetuity, subject to
particulars such as priority, timing, and location. With a
water “lease” as used herein, the holder of the right agrees
to deliver, or allow the buyer access to, a certain quantity of
water over a stated time period, subject to conditions such
as timing of access and location. One-time transfers of
water (essentially short-term leases) are sometimes called
“spot market” trades or “rental” transactions. This paper
reports on both sales and leases of water rights.
METHODS
The broad-scale examination of water prices reported
here is made possible by the Water Strategist and its
predecessor the Water Intelligence Monthly, published
by Stratecon, Inc., of Claremont, California, which have
summarized many of the available western water market
transactions in reports released on a monthly or quarterly
basis. Fourteen years of transactions reported by these
publications (1990-2003) were tabulated to provide the
estimates of the price of water described here. It is
important to note that these publications did not report on
all the transactions that occurred. Especially in the case of
water leases, large numbers of trades were not summarized.
Neither are the included transactions a random sample.
Thus, the current report indicates the nature, but not the
breadth or precise character of western water trades.
Each water transaction entry in the Water Strategist
or Water Intelligence Monthly briefly summarizes one
or more actual trades. The entries do not allow a full
understanding of what influenced the price, and are not
always consistent in how the transactions are described
(perhaps because some information was not available).
Nevertheless, most of the entries provide sufficient
information for a rudimentary analysis of the factors
influencing water market prices, and together they form
the most comprehensive set of information available about
water market trades in the western United States.
The entries typically include buyer, seller, purpose
for which the water was purchased, type of transaction
(whether purchase or lease of a water right), and the source
of the water (surface water, ground water, effluent, or
potable water). Buyers and sellers are categorized herein as
one of the following: (1) municipality; (2) irrigator (farmer
or rancher); (3) private environmental protection entity
(e.g., public trust concern, or private entity such as the
Nature Conservancy); (4) private entity providing water
to many users, such as a water “district,” “association,” or
“company,” referred to here as a “water district”; (5) public
agency (federal or state government agency, conservancy
district, or other water “authority”); (6) other entity (e.g.,
power company, mining company, developer, investor,
country club, feedlot, individual homeowner); or (7)
several entities (when several buyers or sellers of different
types were listed, such that the transaction could not be
neatly assigned to one of the other categories.
The purpose of the transaction was characterized as one
of the following: (1) municipal or domestic (including
commercial and industrial if serviced by a municipality,
and golf courses and other landscape irrigation); (2)
agricultural irrigation; (3) environmental (e.g., instream
flow augmentation); (4) other (e.g., thermoelectric cooling,
recreation, mining, aquifer recharge, augmentation of flows
leaving the state per court order, supply to individual
businesses such as feedlot or manufacturing plant, an
investment of undefined characteristics, unspecified); or (5)
several (several purposes, such that the transaction could
not be neatly assigned to one of the other categories).
Some entries covered several related transactions. For
example, several sellers or several buyers, or both, may have
been included in the entry. Or several transactions within
the same market may have been listed together in the same
entry. Such entries were broken down into separate cases for
analysis if distinct prices were listed and different categories
BROWN
461
Figure 2. Trends in total number
of acre-feet transferred .
of buyers, sellers, or purposes were involved. After this
disaggregation process, a total of 2,447 transactions were
available for the 1990-2003 period.
The Colorado Big Thompson (CBT) market is the most
active market for water rights in the West, with up to
30 or more purchases per quarter by municipalities alone.
It is also a market about which market information is
readily available. The entries listed 949 CBT trades over
the 14 years. Because the sale price for CBT shares differed
little among trades completed during a given month, and
because the volumes traded were typically small (averaging
40 acre-feet), all CBT transactions of a single purpose
within a given month were tabulated as one case for analysis
in order to avoid having CBT transactions overwhelm the
summary statistics. This aggregation process left a total of
228 CBT cases for the 14-year period, and thus a total of
1,726 cases (2,447 - 721 CBT transactions consolidated)
for analysis.
Of these 1,726 cases, 349 were omitted from further
analysis because key information was missing (such as price
or amount of water transferred), something other than
raw water (i.e., effluent or treated water) was involved, the
price included payment for things other than water (e.g.,
land), or the transaction was not a market sale (e.g., it was
an exchange or a donation). Thus, 1,377 qualifying cases
(1,726 - 349 missing information) were left for analysis.
Figure 2 shows the total water volume by year of the
qualifying cases and the full set of cases.
Prices, expressed on a per acre-foot basis, were adjusted
to year 2003 dollars using the consumer price index.
Prices for water rights were converted to an annual basis
using a 3% interest rate, which is approximately the mean
annual growth rate in real gross domestic product over the
past 20 years in the U.S. Although mean prices are also
reported, this analysis emphasizes median prices, which
more accurately indicate the price of a typical water sale
when the price distributions are skewed.
Prices paid for untreated water often include
reimbursement for water management, including such
services as storage and conveyance, in addition to the
cost of the raw water in the stream. Such prices are
analogous to P2 in Figure 1 given supply at Q3. Because our
primary interest is in the value of streamflow, costs of water
management were not included in the price when such
costs could be separated out. However, storage and delivery
services are so commonly a part of water transactions
that such services were often not even mentioned in the
entries. Most prices reported here probably include some
consideration for the value of water management services.
RESULTS
All results reported here are based on the 1,377
cases meeting the criteria for further analysis explained
above. Figure 3 shows the number of cases by a
convenient geographic breakdown, climatic division
(www.cdc.noaa.gov). Fourteen states have qualifying cases
(all states in Figure 3 except North Dakota, South Dakota,
and Nebraska). Three climatic divisions within these states
have over 75 cases: division 4 in northeast Colorado,
including Denver, Fort Collins, and other cities along the
northern Front Range; division 5 in California, capturing
the southern (San Joaquin River) portion of the Central
Valley and on down to the Bakersfield area; and division
10 at the southern tip of Texas, along the Rio Grande as it
enters the Gulf near Brownsville. Nine climatic divisions
had between 26 and 75 cases-three in California, two in
Texas, and one each in Arizona, Colorado, Idaho, and
Nevada. Thirteen climatic divisions had between 11 and
25 cases, and 43 had from 1 to 10 cases. Another 44
climatic divisions in the 14 states had no cases.
462
MARGINAL ECONOMIC VALUE OF STREAMFLOW
Figure 3. Number of cases meeting
criteria for analysis of market
prices, 1990-2003, by climatic
division (divisions are numbered
independently within each state).
Quantity of Water Sold
A median of 804 acre-feet was transferred per case
(the mean is 17,234 acre-feet). Table 1 lists the volume
transferred by state. Three states (Arizona, California, and
Idaho) account for 75% of the water transferred.
In all years much more water has been transferred via
leases than via rights (Figure 4), which reflects in part
the fact that water transfers are easier to agree upon and
arrange on a temporary than on a permanent basis. The
median lease size over the 14 years is 6000 acre-feet per
Table 1. Western water
market activity and prices
by state, 1990-2003 (both Arizona
leases and rights, price in California
year 2003 dollars per acreColorado
foot per year).
Idaho
Kansas
Montana
New Mexico
Nevada
Oklahoma
Oregon
Texas
Utah
* Cases with a $0 price were
Washington
not included. $0 indicates
Wyoming
rounding of a very low price.
All
case, compared with 110 acre-feet for water rights cases.
There is considerable annual variation in amount of water
transferred for both types of transactions, but no apparent
relation between the two trends (R = 0.13).
Ten percent (141) of the cases involve groundwater, with
the remainder (1,236) being of surface water. However,
only 4% of the water transferred in these trades has been
ground water, as suggested by the fact that the average
water volumes per case are 6,679 acre-feet for groundwater
and 18,438 for surface water.
Number of cases
Volume
(N)
(1000 acre-feet)
86
294
427
64
16
5
59
69
3
43
207
43
25
36
1,377
7,910
8,104
443
2,802
11
15
525
1,299
81
261
1,376
122
563
219
23,731
Mean
51
96
133
15
40
20
77
125
246
31
104
34
70
37
96
Price
Median
Min*
48
66
81
6
48
6
55
106
118
9
28
16
32
40
56
0
0
1
0
13
2
1
6
46
1
7
5
3
3
0
Max
115
1,000
630
251
54
56
607
375
575
302
2,258
165
343
93
2,258
BROWN
463
Figure 4. Trends in total quantity
of water transferred (qualifying cases
only).
Price of Water
A quick look at Table 1 reveals at least three findings
of interest about water prices. First, mean prices exceed
median prices for the complete set of cases (a mean of $96
per acre-foot per year versus a median of $56) and for all
but two states. Second, the range in price is substantial for
each state, with most minimums near $0 and maximums
typically in the $100s (the maximum of $2,258 is for
a lease by a mining company). Clearly, water changes
hands at a variety of prices. Third, the median prices vary
substantially among the states, ranging from below $10 in
Idaho and Oregon to over $80 in Colorado and Nevada
(ignoring Oklahoma and Montana because of their small
numbers of cases). There is no apparent relation between
number of cases and median price. Also of note is that
water trades are much more common in some states (e.g.,
California and Colorado) than others (e.g., Montana and
Oklahoma). Water scarcity no doubt plays some role in
determining the number of trades, but institutional and
legal differences are probably the most important factors
affecting sale frequency among the western states.
To begin to understand the reasons for the range of
median prices, consider Table 2, which summarizes the
sales by type of transaction, either a lease or a perpetual
right. Over all states, the median price for leases ($47
per acre-foot per year) is about two-thirds that of rights
($72) given the 3% interest rate for annualizing prices
of rights. However, the median price of leases exceeds
the median annualized price of rights in most states. The
overall superiority of median water rights prices results
largely from the fact that 56% of the water rights cases
are for Colorado, a state where the median price of water
rights far exceeded the median price of leases. Fully 216
(59%) of the 369 water rights cases for Colorado are of
CBT shares, and another 129 (35%) are for other water
rights along the northern Front Range within or near the
area of the Northern Colorado Water Conservancy District
where CBT shares trade (in climate division 4, Figure 3).
Also of interest is that for 10 of the 14 states the number
of lease cases exceeded the number of rights cases. The
exceptionally high number of water rights transactions
in Colorado reflects the relative ease with which such
transactions can be accomplished and the strong demand
for secure water supplies by the fast-growing cities along
the Front Range.
Table 3 summarizes the cases by the purpose for which
the water was purchased. Over half (739) of the purchases
were for municipal purposes, another 23% (321) were
for irrigation, and 11% (150) were for environmental
purposes. The median price paid for municipal uses ($77)
Table 2. Western water market prices by state and type of
transaction, 1990-2003 (year 2003 dollars per acre-foot per
year).
Leases
Median
N
($)
Arizona
California
Colorado
Idaho
Kansas
Montana
New Mexico
Nevada
Oklahoma
Oregon
Texas
Utah
Washington
Wyoming
All
48
250
58
49
11
5
29
4
2
34
159
11
21
34
715
58
68
18
8
50
6
55
83
347
9
29
7
37
40
47
Rights
Median
N
($)
38
44
369
15
5
0
30
65
1
9
48
32
4
2
662
40
37
84
3
16
76
109
46
7
24
17
13
43
72
464
MARGINAL ECONOMIC VALUE OF STREAMFLOW
Table 3. Western water market prices by purpose of buyer,
1990-2003 (both leases and rights, year 2003 dollars per acrefoot per year).
Purpose
Municipal
Irrigation
Environment
Other
Several
All
N
739
321
150
105
62
1377
Mean Median
($)
($)
118
46
56
180
51
96
77
28
40
62
56
56
Min
($)#
Max
($)
0
1
0
2
2
0
1607
490
450
2258
190
2258
# Cases with a $0 price were not included. $0 indicates rounding
of a very low price.
was nearly three times that paid for irrigation water ($28)
and nearly twice that paid for environmental purposes
($40). Purchases for municipal purposes tended to be of
water rights (453 cases involving rights versus 286 for
leases), suggesting that municipalities desire-and are able
to pay for-dependability of supply. Purchases for irrigation
and environmental purposes tended to be of leases.
For the three principal purposes for which water was
purchased (municipal supply, irrigation, and environmental
protection), there is a wide range in median price across
the states (Table 4). The overall median price paid for
municipal water ($77 per acre-foot per year) is heavily
influenced by sales in Colorado, where the median price of
the 250 cases is $88. Other states with both high median
Table 4. Western water market prices by state and purpose of
buyer, 1990-2003 (both leases and rights, year 2003 dollars per
acre-foot per year).
Municipal
Median
N
($)
Arizona
California
Colorado
Idaho
Kansas
Montana
New Mexico
Nevada
Oklahoma
Oregon
Texas
Utah
Washington
Wyoming
All
47
149
250
5
14
0
26
59
3
0
141
28
3
14
739
48
96
88
6
48
77
110
118
26
20
40
77
77
Irrigation
Environmental
Median
Median
N
($)
N
($)
12
66
110
31
2
1
4
0
0
20
40
13
5
17
321
45
45
72
4
51
6
50
8
24
7
17
5
28
5
51
19
19
0
3
10
7
0
19
0
2
15
0
150
45
64
20
8
2
47
43
26
40
32
40
prices for municipal purposes and a substantial number of
cases are California (median of $96), Nevada ($110), and
Wyoming ($77). Excepting Colorado, the median price
of the remaining 489 sales for municipal purposes is $57.
The overall median price paid for irrigation water ($28)
is also heavily influenced by Colorado, which had over
one-third of these cases and a median price of $72.
Other states with a substantial number of cases include
California (median of $45), Idaho ($4), and Texas ($24).
Excepting Colorado, the median price of the remaining
211 cases for irrigation purposes is $16. Among purchases
for environmental purposes, states with the highest
median prices and with at least ten cases are California
($64), Colorado ($20), Idaho ($8), Oregon ($26), and
Washington ($32).
Who Sold to Whom, and for What Purpose?
Irrigators are the sellers in 38% (531) of the cases (Table
5), not counting when they might appear among the 222
cases involving several sellers. Public agencies are the sellers
for another 16% (224) of the cases. These public agency
sales include those involving State Water Project or Central
Valley Project water in California, Central Arizona Project
water in Arizona, and water managed by the U.S. Bureau of
Reclamation in many states, including Colorado, Oregon,
and Wyoming. Nearly all public agency sales were leases.
Municipalities were the most common buyers of water,
accounting for 27% (375) of the cases (Table 5), not
counting when they might appear among the 206 cases
involving several buyers. Other active buyers were public
agencies with 17% (234) of the cases, farmers with 15%
(204) of the cases, and water districts with 13% (183) of
the cases.
Trends in Occurrence and Price
The number of cases per year (across all states and
purposes) ranges from a minimum of 77 in 1995 to a
maximum of 142 in 1999 (Figure 5). Recent years show an
increase in the number of cases (over the 14 year period,
the four highest numbers of cases occurred in 1999, 2001,
2002, and 2003); the overall increasing trend is statistically
significant at the 0.05 probability level based on the MannKendall test for time trends (test statistic = k = 2.56).
Examining sales of leases and rights separately reveals that
the number of leases has increased substantially over the
past 14 years (k = 3.60), whereas the numbers of sales of
rights show no trend (k = -0.49).
The median price per year (across all states and purposes)
ranges from $34 in 1998 to $76 in 1991 (Figure 5).
No overall trend is evident (k = 0.24). However, looking
BROWN
465
Table 5. Number of western water market trades from seller to buyer, 1990-2003 (both leases and rights).
Seller
Buyer
Municipality Irrigation Environmental
Municipality
Irrigator
Environmental
Water district
Public agency
Other
Several
Total
26
175
1
21
39
60
53
375
8
96
0
19
48
12
21
204
0
10
1
1
1
1
0
14
Total
Water
district
Public
agency
Other
Several
7
60
0
38
32
26
20
183
18
95
1
32
46
22
20
234
6
40
0
8
18
68
21
161
4
55
0
16
40
4
87
206
69
531
3
135
224
193
222
1377
Figure 5. Trend in median
price of water, all water uses
(includes both leases and rights,
year 2003 dollars).
separately at sales of leases and rights reveals that the
median price of rights has increased significantly (k = 2.07),
whereas the price of leases has not (k = 0.00). Colorado,
especially the Front Range area, is largely responsible for
the overall increase in the price of water rights.
Price Differences Across Markets
Space constraints preclude presenting details here about
individual water markets. An analysis of separate markets
revealed the following findings. First, prices at a given
time can vary considerably even among markets located
quite close to each other. Such markets often differ
in local economic conditions, availability of alternative
supplies (such as groundwater as a supplement for surface
water), extent of water distribution infrastructure, and past
decisions to obtain secure surface water rights. Second,
prices in competitive markets can change dramatically over
time in response to development pressures or weather
cycles. Third, prices of leases in many markets are
heavily influenced by administrative criteria, and thus not
competitively set (exceptions to this observation include the
Texas Rio Grande lease market). Fourth, water rights are
typically sold in relatively competitive situations where the
prices are determined by individual negotiations between
buyer and seller.
CONCLUSIONS
Analysis of the trades reported by Stratecon allows the
following general statements (which may not represent the
full population of western trades):
1. The incidence of water market trades is geographically
variable. Markets are very active in a few areas of the West,
but most areas apparently had few trades over the past
14 years. Although three states (California, Colorado, and
Texas) account for three-fourths of the qualifying sales,
even in these states some areas had very few trades.
2. In a given year, at least ten times as much water
changes hands via leases as changes hands via sales of water
rights. The median size of leases is over 50 times that of
water rights sales.
466
MARGINAL ECONOMIC VALUE OF STREAMFLOW
3. Across the western states, the median price of water
is highly variable, with Colorado and Nevada having the
highest medians, and Idaho and Oregon having the lowest
medians, when sales of leases and rights are combined.
However, the price of water is also highly variable within
every state, reflecting the particular physical and legal
characteristics of individual water markets. This variability
makes it risky to transfer a value from one location to
another, thus complicating the process of benefit transfer.
4. Among the major purposes for which water was
purchased, purchases for municipal uses have the highest
median price ($77) and account for over half of all trades.
Purchases for agricultural irrigation and environmental
protection have lower median prices (roughly $35).
5. Purchases for municipal purposes have tended to
be of water rights, whereas purchases for irrigation,
environmental, or other purposes have tended to be of
leases.
6. Irrigators were the sellers in about 40% of the
transactions. Public agencies, such as federal agencies
managing large water storage and delivery projects, were
the sellers in another 16% or so of the transactions.
Municipalities were the most common buyers, accounting
for about 30% of the transactions. Other common buyers
were farmers, public agencies, and water districts.
7. Across all cases, the median price of leases in real terms
showed no consistent trend over the past 14 years, whereas
the median price of water rights showed an upward trend.
Water market activity in aggregate offers a broad and
rich understanding of the value of water. Water values can
be substantial, but because they also are highly variable
both geographically and temporarily, care must be used in
applying water market prices to analyze policies affecting
streamflow.
Acknowledgement: Alex Bujak, research assistant with
Colorado State University, ably helped summarize the
transactions and maintain the database.
LITERATURE CITED
Brown, TC. 2006. Trends in water market activity and price
in the western United States. Water Resources Research 42,
WO9402, doi: 10.1029/2005WR004180.
Brown, TC, BL Harding, and EA Payton. 1990. Marginal
economic value of streamflow: A case study for the Colorado
River Basin. Water Resources Research 26(12): 2845-2859.
Eheart, JW, and RM Lyon. 1983. Alternative structures for water
rights markets. Water Resources Research 19(4): 887-894.
Gibbons, DC. 1986. The economic value of water. Washington,
D.C.: Resources for the Future.
Gillilan, DM, and TC Brown. 1997. Instream flow protection:
Seeking a balance in western water use. Washington, DC:
Island Press.
Hartman, LM, and D Seastone. 1970. Water transfers: Economic
efficiency and alternative institutions. Baltimore, MD: Johns
Hopkins Press.
Howe, CW, and C Goemans. 2003. Water transfers and their
impacts: Lessons from three Colorado water markets. Journal
of the American Water Resources Association 39(5): 1055-1065.
Howe, CW, DR Schurmeier, and WDJ Shaw. 1986. Innovations
in water management: Lessons from the Colorado-Big
Thompson Project and Northern Colorado Water
Conservancy District. In: KD Frederick (ed.), Scarce water
and institutional change. Washington, D.C.: Resources for
the Future: 171-200.
Loomis, JB. 1992. The 1991 State of California Water Bank:
Water marketing takes a quantum leap. Rivers 3(2): 129-134.
Michelsen, AM. 1994. Administrative, institutional, and
structural characteristics of an active water market. Water
Resources Bulletin 30(6): 971-982.
National Research Council. 1992. Water transfers in the West:
efficiency, equity, and the environment. Washington, DC:
National Academy Press.
North, DC. 1990. Institutions, institutional change, and
economic performance. New York: Cambridge University
Press.
Saliba, BC. 1987. Do water markets “work”? Market transfers
and trade-offs in the southwestern states. Water Resources
Research 23(7): 1113-1122.
Saliba, BC, DB Bush, and WE Martin. 1987. Water marketing
in the Southwest--can market prices be used to evaluate water
supply augmentation projects?, General Technical Report
RM-144. Fort Collins, CO: USDA Forest Service, Rocky
Mountain Forest and Range Experiment Station,.
Young, RA. 1996. Measuring economic benefits for water
investments and policies, Technical Paper 338. Washington,
DC: World Bank.